基于模型的诊断推理

给出计算基于模型诊断的能用算法;通过对带任意初始条件的测试刻画鉴别诊断来缩小该诊峡占?通过给出高效的生成分级模型以及基于分级模型的将诊断产生与测试产生相结合的算法来减少计算的复杂性;将时态推理、不确定性推理等方法有机地结合到设备诊断中去；在锒瞎讨锌悸俏薏⑹褂米远婊燃际跄甭砸宰钚〉拇凼股璞富指凑！

Model-based Diagnosis is a new intelligent diagnostic technique proposed to overcome the serious drawbacks of traditional diagnostic expert systems. In this project, a general algorithm for computing model-based diagnosis is given. The initial test condition of the Differential Diagnosis Principles is generalized, which improves the adaptability of differential diagnosis test and helps reduce the diagnostic space. Hierarchical diagnosis that can help reduce the complexity of model-based diagnosis is researched into. The soundness of hierarchical diagnosis is proved and the incompleteness of hierarchical diagnosis is pointed out. The concepts of model-based diagnosis and kernel model-based diagnosis are given when the model of the system to be diagnosed is uncertain causal theory. The direct relationship between kernel model-based diagnosis and U-S-prime implicants/implicates is demonstrated. The concepts of component replacing and replacement test are proposed and the discrimination of diagnoses and the determination of faulty components are explored. These results can help select the components to be replaced, improve the effectiveness of diagnosing, and restore the normal functions of the system being diagnosed as quickly as possible by combining testing with repairing.